Preparing MPICH for exascale
- 9 January 2025
- journal article
- research article
- Published by SAGE Publications in The International Journal of High Performance Computing Applications
- Vol. 39 (2) , 283-305
Abstract
The advent of exascale supercomputers heralds a new era of scientific discovery, yet it introduces significant architectural challenges that must be overcome for MPI applications to fully exploit its potential. Among these challenges is the adoption of heterogeneous architectures, particularly the integration of GPUs to accelerate computation. Additionally, the complexity of multithreaded programming models has also become a critical factor in achieving performance at scale. The efficient utilization of hardware acceleration for communication, provided by modern NICs, is also essential for achieving low latency and high throughput communication in such complex systems. In response to these challenges, the MPICH library, a high-performance and widely used Message Passing Interface (MPI) implementation, has undergone significant enhancements. This paper presents four major contributions that prepare MPICH for the exascale transition. First, we describe a lightweight communication stack that leverages the advanced features of modern NICs to maximize hardware acceleration. Second, our work showcases a highly scalable multithreaded communication model that addresses the complexities of concurrent environments. Third, we introduce GPU-aware communication capabilities that optimize data movement in GPU-integrated systems. Finally, we present a new datatype engine aimed at accelerating the use of MPI derived datatypes on GPUs. These improvements in the MPICH library not only address the immediate needs of exascale computing architectures but also set a foundation for exploiting future innovations in high-performance computing. By embracing these new designs and approaches, MPICH-derived libraries from HPE Cray and Intel were able to achieve real exascale performance on OLCF Frontier and ALCF Aurora respectively.Keywords
Funding Information
- Exascale Computing Project (17-SC-20-SC)
- U.S. Department of Energy, Office of Science (DE-AC02-06CH11357)
- U.S. DOE Office of Science Advanced Scientific Computing Research Program (DE-AC02-06CH11357)
- U.S. Department of Energy (DE-AC05-00OR22725)
This publication has 20 references indexed in Scilit:
- Network Assisted Non-Contiguous Transfers for GPU-Aware MPI Libraries2022
- FALCON-X: Zero-copy MPI derived datatype processing on modern CPU and GPU architecturesJournal of Parallel and Distributed Computing, 2020
- A survey of MPI usage in the US exascale computing projectConcurrency and Computation: Practice and Experience, 2018
- QMCPACK: an open sourceab initioquantum Monte Carlo package for the electronic structure of atoms, molecules and solidsJournal of Physics: Condensed Matter, 2018
- CUDA Kernel Based Collective Reduction Operations on Large-scale GPU Clusters2016
- Assessing a mini‐application as a performance proxy for a finite element method engineering applicationConcurrency and Computation: Practice and Experience, 2015
- NWChem: A comprehensive and scalable open-source solution for large scale molecular simulationsComputer Physics Communications, 2010
- PMI: A Scalable Parallel Process-Management Interface for Extreme-Scale Systems2010
- Implementing Fast and Reusable Datatype Processing2003
- The Locally Self-consistent Multiple Scattering code in a geographically distributed linked MPP environmentParallel Computing, 1998